NMDA receptors (NMDARs) are glutamate receptor ion channels that mediate excitatory neurotransmission. The majority of NMDARs are composed of two GluN1 and two GluN2 subunits. One GluN1 subunit has been cloned, but there are four GluN2 subunits (GluN2A-D) that endow NMDARs with distinct functional properties and different developmental and regional expression. Mutations in the gene encoding the GluN2A subunit have been associated with childhood epilepsy/aphasia syndromes, and a subset of these mutations cause gain-of- function in GluN2A-containing receptors that lead to severe neurologic complications. The high frequency of GluN2A mutations linked to neurologic conditions provides a compelling rationale for the development of GluN2A-selective therapeutic agents. GluN2A-selective modulators can also be useful and much needed pharmacological tools for neurophysiological studies. We have identified novel GluN2A-selective negative allosteric modulators (NAMs) that inhibit NMDARs by reducing glycine binding to the GluN1 subunit. These NAMs can transform our ability to evaluate the contribution of GluN2A to normal brain function and disease. In addition, we have determined crystal structures of the heterodimer formed by GluN1 and GluN2A agonist binding domains (ABDs) with NAMs bound at the subunit interface. These crystal structures represent the discovery of a novel, unexplored modulatory site and provide new opportunities for the development of subunit- selective ligands. We will uncover structural and mechanistic features of this modulatory site and provide important understanding required for the use of GluN2A-selective NAMs as pharmacological tools. Aim 1) What are the structural bases for selectivity, potency, and efficacy of GluN2A-selective NAMs? We will use crystallography, mutagenesis, and electrophysiological recordings of NMDA receptor function to define the structural determinants for NAM inhibition. Aim 2) What is the mechanistic basis for allosteric interaction between NAM and glycine binding sites? Crystallographic, pharmacological, and mutational analyses will dissect the conformational changes and mechanism that causes the allosteric interaction between NAM and glycine binding sites. Aim 3) Can GluN2A-selective NAMs inhibit triheteromeric and neuronal NMDA receptors? We will define NAM inhibition of NMDARs in conditions relevant to synaptic transmission as well as NAM inhibition of NMDARs that contain gain-of-function GluN2A mutations identified in epilepsy patients. Neuronal NMDARs are heterogeneous populations of diheteromeric receptors comprised of two GluN1 and two identical GluN2 subunits (e.g. GluN1/GluN2A) as well as triheteromeric receptors containing two GluN1 and two different GluN2 subunits (e.g. GluN1/GluN2A/GluN2B). Evaluation of GluN2A-selective modulators on both diheteromeric and triheteromeric receptors is required for their use as pharmacological tools. In addition, we will inhibit NMDAR-mediated synaptic currents in primary cultures of autaptic neurons that express different ratios of GluN2A and GluN2B as well as autaptic neurons that express GluN2A with gain-of-function mutations.